Project #1 seeks to investigate the quantitative retinal output of ganglion cells subserving color vision in the ground squirrel retina. Classically defined 'non-opponent' or luminosity units which nevertheless show quantitative wavelength sensitivity in their output, will be tested for apparent 'color-shifts' due to intensity (Bezold-Brucke effect) as will opponent color-coded units. Gaussian white noise analysis will be made in an attempt to provide a quantitative signature for the various types of units investigated. Project #2 seeks to extend systems analysis (both linear and nonlinear) to continue characterizing the response properties of neurons that fire repetitively. The study will concentrate on extending observations made on the tonic stretch receptor of the crayfish and on the eccentric cell of Limulus to the phasic stretch receptor of the crayfish. Consequences due to the nonlinearity of encoder output will be examined with respect to its interaction with feedback and its significance to information processing.